Better indexes

Subject: Better indexes
From: geoff-h -at- MTL -dot- FERIC -dot- CA
Date: Thu, 1 May 1997 12:28:48 -0500

Dennis Hays wrote <<[WordPerfect] had a much superior
method. You could create a "concordance" file--a
separate file with just a list of index words. WordPerfect
would compare this file against the doc file... and build
the index.>>

I can hear Lori Lathrop's howls already! <grin> Having used
WordPerfect's documentation in earlier versions, I have to
say that they produced some of the least helpful indexes
I'd ever used. Dennis is right about the efficiency of this
approach, but it's important to remember that a concordance
is not an index... it's the starting point for building
one, and arguably not the most efficient starting point. If
you stop indexing at the concordance stage, all you get is
a list of words stripped of context, and readers need
context to know whether to flip to page 17 or page 233 to
find out what the reference "printer" means. An index will
tell you "printer: configuring" is on page 233 and
"printer: purchasing" is on page 17.

The problem with indexing is simple, but nasty: you can't
automate the job because you need a human to define the
context for each reference, and to do so in the context of
the reader's needs. Until the technology gets much, much
better, indexing is still a human job.

--Geoff Hart @8^{)} geoff-h -at- mtl -dot- feric -dot- ca
Disclaimer: Speaking for myself, not FERIC.

TECHWR-L (Technical Communication) List Information: To send a message
to 2500+ readers, e-mail to TECHWR-L -at- LISTSERV -dot- OKSTATE -dot- EDU -dot- Send commands
Search the archives at or search and
browse the archives at

Previous by Author: Information mapping & Author-developers
Next by Author: Developers do docs/turnaround time
Previous by Thread: WARNING - ISP hustle
Next by Thread: Concordance and Indexing

What this post helpful? Share it with friends and colleagues:

Sponsored Ads